Teardown and Repair: 20W Tubelight LED Driver

Tubelight Internals

Tubelight at home randomly stopped working this week for no apparent reason. Opened it up to see what went wrong. Started probing and saw that the input fuse (Supposedly a fuse but they put a 61Ohm resistor) was blown. Replaced that and the 41Ω, 1W resistor on the return path, and the tube light was back to normal. Not sure what caused the blow-up. Maybe the resistor degraded with time.

20W Tubelight Schematic and circuit
20W Tubelight Schematic and circuit

I went through and traced out the circuit for anyone wanting to know its schematic. The AC Input side is current limited with a “fuse” and fed into a classic full wave bridge rectifier(MB105) IC to convert to high voltage DC. There is space given in the input for filtration/safety capacitors but they have not populated it. Probably to save a few cents, might have put it there to pass the testing. The output of the rectifier contains a CY400 capacitor and Metal Oxide Varistors(MOV). This is fed to SIC9762 which is a high voltage LED driver with an integrated MOSFET. It’s a constant current LED driver with the drive current set at around 200mA with a sense resistor. On the output LED side, you have a large electrolytic capacitor of 100uF with a bleeding resistor to discharge the tube light immediately(else you will get a ghosting effect of the LED). There are 100 LEDs in total at the output.

All in all, it’s a penny-pinched circuit optimized to minimize the BOM to the lowest price possible. That’s exactly the reason you get these tube lights in the $3-$4 range per unit in the retail market.

If you liked the post, Share it with your friends!

Rant: The USB Type C Cable Chaos

USB Type C Breakout Board

Recently, I misplaced my OnePlus phone’s charging cable and resorted to a Type A to Type C USB3 cable with their original charger. Voila, OnePlus’s superfast charging didn’t kick in with the alternate cable, even though it was of better quality than their official one. This is not a new dilemma; I was curious about how they detect the cable change. Looking at the Type A end of the cable, it appeared similar to a regular USB2.0 Type A connector with the standard four lines: VBus, D+, D-, GND. Interestingly, there was an exposed middle GND-Drain line, which seemed odd. My hunch was that they might be employing some non-standard 1-wire communication over that line.

To delve deeper, I attempted to use a USB TypeC Breakout and probed each pin on the Type C end, but the GND-Drain pin wasn’t connected anywhere. This strongly hinted at the use of an eMarker chip in the cable head. These minuscule chips play a pivotal role in communication between devices and chargers, and that’s where the conundrum arises.

USB Type C cables were intended to establish uniformity and standards, but the industry has managed to turn it into a colossal mess. The market is flooded with seemingly identical Type C cables that differ drastically in performance. And that’s just the physical design part – add to it the proprietary charging protocols introduced by each company. From Apple to Samsung, Qualcomm to Xiaomi, and OnePlus/Oppo, every brand is pushing its unique charging approach, accompanied by flashy names like QuickCharge, Dash/Warp Charge, SuperCharge, and PD fast charging. Just to figure out the Voltage and Current needed.

What’s preventing companies from adhering to a universal standard? The lucrative $50+ billion cable and accessories market. If you purchase a device, you better buy its specific cables or risk a drop in performance due to vendor lock-ins. This issue becomes even more vexing today as companies often exclude chargers and cables when you buy a phone.

This mess won’t be sorted out anytime soon, with USB4 introducing more confusion. Wired cables will be weird in the years to come.

If you liked the post, Share it with your friends!
1 2 3 4 20